[SciPy-user] installation on Suse8.1 (blas problem)

[Pearu Peterson]

On Mon, 17 Mar 2003 baecker at physik.tu-dresden.de wrote:

> 
> 
> On Mon, 17 Mar 2003, Pearu Peterson wrote:
> 
> [...]
> > Here I get
> >
> > >>> scipy.linalg.fblas.dnrm2([3,-4,5])
> > 7.0710678118654755
> > >>> scipy.linalg.fblas.snrm2([3,-4,5])
> > 7.0710678100585938
> >
> 
> on a different machine I can confirm these numbers till the last digit ...

That's normal for Python float.__repr__.

> > What Fortran compiler flags were used to compile the blas library?
> > Note that using 3dnow instructions may cause differences in floating
> > point results but I never thought it would be so noticable.
> 
> well, I just use the default ones from scipy (i.e. no change).
> How can I figure out the ones which are used?

scipy defaults should not affect the behaviour of prebuilt Fortran
compiled libraries. Anyway, to find out what Fortran compiler flags are
used when building scipy, execute

>>> from scipy_distutils.command.build_flib import find_fortran_compiler
>>> print find_fortran_compiler(None,None,None,verbose=0)

OTOH, my question may be irrelevant since you are using blas library
provided by Suse... you can try to build blas libraries yourself.

Pearu